Premium
Copy number variation analysis and targeted NGS in 77 families with suspected Lynch syndrome reveals novel potential causative genes
Author(s) -
Kayser Katrin,
Degenhardt Franziska,
Holzapfel Stefanie,
Horpaopan Sukanya,
Peters Sophia,
Spier Isabel,
Morak Monika,
Vangala Deepak,
Rahner Nils,
KnebelDoeberitz Magnus,
Schackert Hans K.,
Engel Christoph,
Büttner Reinhard,
Wijnen Juul,
Doerks Tobias,
Bork Peer,
Moebus Susanne,
Herms Stefan,
Fischer Sascha,
Hoffmann Per,
Aretz Stefan,
SteinkeLange Verena
Publication year - 2018
Publication title -
international journal of cancer
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.475
H-Index - 234
eISSN - 1097-0215
pISSN - 0020-7136
DOI - 10.1002/ijc.31725
Subject(s) - copy number variation , genetics , sanger sequencing , biology , candidate gene , lynch syndrome , gene , copy number analysis , msh2 , mutation , dna mismatch repair , genome , dna repair
In many families with suspected Lynch syndrome (LS), no germline mutation in the causative mismatch repair (MMR) genes is detected during routine diagnostics. To identify novel causative genes for LS, the present study investigated 77 unrelated, mutation‐negative patients with clinically suspected LS and a loss of MSH2 in tumor tissue. An analysis for genomic copy number variants (CNV) was performed, with subsequent next generation sequencing (NGS) of selected candidate genes in a subgroup of the cohort. Genomic DNA was genotyped using Illumina's HumanOmniExpress Bead Array. After quality control and filtering, 25 deletions and 16 duplications encompassing 73 genes were identified in 28 patients. No recurrent CNV was detected, and none of the CNVs affected the regulatory regions of MSH2 . A total of 49 candidate genes from genomic regions implicated by the present CNV analysis and 30 known or assumed risk genes for colorectal cancer (CRC) were then sequenced in a subset of 38 patients using a customized NGS gene panel and Sanger sequencing. Single nucleotide variants were identified in 14 candidate genes from the CNV analysis. The most promising of these candidate genes were: ( i ) PRKCA , PRKDC , and MCM4, as a functional relation to MSH2 is predicted by network analysis, and ( ii ) CSMD1 , as this is commonly mutated in CRC. Furthermore, six patients harbored POLE variants outside the exonuclease domain, suggesting that these might be implicated in hereditary CRC. Analyses in larger cohorts of suspected LS patients recruited via international collaborations are warranted to verify the present findings.